| | |
Summary: Gaussian Energy Functions for Registration without Correspondences
Faysal Boughorbel, Andreas Koschan, Besma Abidi, and Mongi Abidi
Imaging Robotics and Intelligent Systems Laboratory
The University of Tennessee, Knoxville, TN, 37996
fboughor@utk.edu
Abstract
A new criterion based on Gaussian fields is introduced
and applied to the task of automatic rigid registration of
point-sets. The method defines a simple energy function,
which is always differentiable and convex in a large
neighborhood of the alignment parameters; allowing for
the use of powerful standard optimization techniques. We
show that the size of the region of convergence can be
extended so that no close initialization is needed, thus
overcoming local convergence problems of Iterative
Closest Point algorithms. Furthermore, the Gaussian
energy function can be evaluated with linear complexity
using the Fast Gauss Transform, which permits efficient
implementation of the registration algorithm. Analysis
through several experimental results on real world
|
